The power of the solar rays depends on the manner in which they fall, as may be seen from the difference of climates. The earth is about 3,000,000 of miles nearer to the sun in winter than in summer, but the rays strike the northern hemisphere more obliquely in winter than in the other half of the year. Diurnal variations of heat are perceptible only to a small distance below the surface of the ground, because the earth is a bad conductor: the annual influence of the sun penetrates much farther. At the equator, where the heat is greatest, it descends deeper than elsewhere with a diminishing intensity, but there, and everywhere throughout the globe, there is a stratum, at a depth varying from 40 to 100 feet below the surface of the ground, where the temperature never varies, and is nearly the same with the mean temperature of the country over it. This zone, unaffected by the sun’s heat from above, or by the internal heat from below, serves as an origin whence the effects of solar heat are estimated on one hand, and the internal temperature of the globe on the other. Below it the heat of the earth increases, as already mentioned, at the rate of one degree of Fahrenheit’s thermometer for every 50 or 60 feet of perpendicular depth; were it to continue increasing at that rate, every substance would be in a state of fusion at the depth of 21 miles; hitherto, however, the experiments in mines and Artesian wells, whence the earth’s temperature below the constant stratum is ascertained, have not been extended below 1700 feet. M. de Beaumont has estimated by the theory of Fourier, from the observations of M. Arago, that the quantity of central heat which reaches the surface of the earth is capable, in the course of a year, of melting a shell of ice covering the globe a quarter of an inch thick.[127] The superficial temperature of the earth is great at the equator, it decreases gradually towards the poles, and is an exact mean between the two at the 45th parallel of latitude; but a multitude of causes disturb this law even between the tropics. It is affected chiefly by the unequal distribution of land and water, by the height above the sea, by the nature of the soil, and by vegetation, so that a line drawn on a map through all the places where the mean temperature of the earth is the same would be very far from coinciding with the parallels of latitude, but would approximate more to them near the equator. Between the tropics the temperature of the earth’s surface is greater in the interior of continents than on the sea-coasts and islands, and in the interior of Africa it is greater than in any other part of the globe.